> ## Content Index
> Fetch the complete content index at: https://www.superjumpmagazine.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# A Thousand Faces, A Thousand Logs
- URL: https://www.superjumpmagazine.com/a-thousand-logs-a-thousand-faces/
- Published: 2026-10-06T07:06:43.000Z
- Updated: 2026-10-06T20:06:23.000Z
- Description: On the meaning of video game interaction and its emerging collision with generative AI
- Author: Shun Takata
- Tags: Game Design, Opinion, AI, Super Mario Bros., Super Mario, Nintendo, Famicom, Serial Experiments Lain, PlayStation

## Prologue: The Girl Who Was Made of Logs

**I am sitting in a dark room before a CRT television, enveloped in faint static. I ask myself what, exactly, I'm doing here.**

The warm glow of the screen emits a complex image from Serial Experiments Lain, released in 1998\. It's filled with faded and unsettling files, fragments of someone's murmurs, clinical audio logs, and images that seem to intersect without context. There's no exhilaration from fast-moving objects and there are no enemies to shoot for instant gratification. Very little is demanded of my fingers. Instead, I scroll. I play back. I pick my way through the remnants of information scattered across the screen, searching for connections between the fragments.

While my fingers have barely moved, a region of my brain has begun heating up; slow fatigue settles in accompanied by a peculiar sense that I've *done something real*. Was I simply a passive spectator, receiving information poured forth from the fuzzy CRT? Or, as I wandered through that sea of fragmented information - questioning its contradictions, searching for connections, and struggling to piece it all together - was there actually a two-way dialogue taking place? Is *moving my body* the only form of interactivity I can experience with digital entertainment?

What if the world can shake us with an unsettling question from the other side of the screen, forcing us to redraw the boundaries of our own understanding?

We receive information from the world. We interpret it. We doubt it. And then we return to search again.

Can we really not call that experience "interactive": the feeling of a world entering into us?

We are entering an age where technology is beginning to anticipate human thought; where *unlimited convenience* means *absence of friction*. Standing at the threshold of this age, I contemplate the very meaning of interactivity. 

What makes an experience interactive?

## Chapter 1: The Hero with a Thousand Faces

I slide the Super Mario Bros. cartridge into the NES, pick up the controller, press Start, and push right on the D-Pad. The small on-screen figure in red begins to run, responding directly to the movement of my finger.

This is Mario. Who is Mario?

We tend to think of characters in games as either hollow puppets we manipulate or independent people who we direct as a kind of digital partner. Mario does not fit neatly into either category.

His simple silhouette. His lack of dialogue. His stylised movements. The clear physical response he gives to every input from the player. Taken together, these qualities establish a kind of space into which I can project my intentions.

Mario is therefore neither a hollow puppet nor an independent collaborator: for the minutes (or hours) of time I spend with him, I am *borrowing his body*.

We press a button. Mario runs. He jumps, breaks blocks, flies through the air, lands under the pull of gravity.

![](https://storage.ghost.io/c/c0/e7/c0e78dd7-e7af-4603-8ab9-09f05cab06f3/content/images/2026/10/super-mario-bros_0-1.png)

Mario's movement is a direct extension of our own. Source: Screen Rant.

The physical movements of my hands are transmitted through the controller prompting changes on the screen leading to further physical responses. I act upon the world - **push** \- and the world responds - **pull**. If I don't move Mario, he doesn't simply decide to run on his own. And if I hesitate just before a jump, he'll fall into a pit. The digital world waits for my intervention. And when I intervene, that world returns a result according to its internal rules.

In biology, the term *bilateral symmetry* means: 

*A body plan where an organism can be divided into matching left and right *mirror-image halves* along a single central plane.*

My relationship with Mario and his Mushroom Kingdom is also a form of bilateral symmetry. It represents a stunningly simple, yet remarkably powerful, dialogue between the video game and the human being playing it.

This is a moment where I experience, in a very physical sense, the philosophy I once described as "[It's You](https://www.superjumpmagazine.com/the-100-yen-democracy-part-ii-the-hero-with-a-thousand-faces/)".

The title of this chapter - and this essay - references the idea of "A Thousand Faces". But this doesn't mean that Mario himself has a thousand different personalities; quite the opposite, in fact. There is a body called Mario: solid and distinct enough to traverse his digital kingdom, yet open enough to welcome the intentions of millions of players worldwide who project their intentions onto him. Mario is the other side of the mirror; the hero we "wear", the inviting reflection who subtly changes for every single human subject connected to him through a controller.

Action, response. Response, action. This is the endless cycle of interactivity that so many of us learned by growing up with video games.

?

The world is waiting.

Press to Jump Inputs: **0** 

InputResponseInterpretationNew Input 

## Chapter 2: Embodied Feedback

Pressing a button and having an interactive experience might seem similar, but they're fundamentally different. When we press a button on a controller, the satisfaction we feel does not come from the fact that the plastic button itself has sunk beneath our finger. Nor does it come solely from seeing pixels rendered in 4K.

At the heart of what we experience is something more like this cycle:

**Input → Response → Interpretation → New Input**

We press a button and the on-screen character jumps. We perceive the way the character lands, the distance between our digital avatar and an enemy, and the many changes taking place on screen through sight, sound, and touch. We interpret these patterns continuously and almost instantaneously.

"Should I have jumped a little earlier?"

"Next time, I should move forward sooner to avoid that enemy."

The way we interpret the game's response and the judgments we make based on said interpretation feed into the next input. Over time, this cycle becomes something far greater than a simple operation/reaction loop: something far more profound emerges.

*A dialogue emerges.*

Throughout this story, I will name this cycle of response mediated through the body as **embodied feedback**.

What matters here is that neither the *amount of information* nor the *resolution level* guarantee, in and of themselves, the density of this dialogue. Even a world rendered in photorealistic detail can fail to achieve a flowing dialogue with the player (if, for example, the player's actions merely advance completely predetermined sequences, leaving little room to read the consequences of an input and construct the next action).

By contrast, a game as simple as Super Mario Bros. - even with its low-fidelity graphics by today's standards - asked us to read the consequences of even the tiniest differences in our inputs. Super Mario Bros. establishes a rich and dynamic dialogue with players for this reason; it doesn't require the latest cutting-edge graphics technology to achieve this.

Throughout the history of digital media, as hardware technology evolved, it was often assumed that high resolution and high interactivity were somehow synonymous: that they naturally evolved in lockstep together. If you can *press more buttons* or *render more detailed graphics*, then surely your experience becomes *more interactive* and the dialogue between game and player becomes all the richer. I'm sure you can think of a game you've played that checked all the right boxes on paper but still somehow felt listless from an interactivity perspective.

Given this, it simply can't be the case that being *interactive* necessarily means *keeping our fingers busy*. And if that's the conclusion we can draw from the discussion, at least so far, then it begs another important question: What if there are forms of media that require *almost no physical manipulation* but that still produce the kind of interactive dialogue I've been articulating so far? These are forms of media that still demand something of us and give us something in return. 

But if we aren't pressing lots of buttons to produce nuanced on-screen action, then in what sense might these kinds of worlds *respond to us*?

Records — open in any order

Lda\_118her diary Tda\_046the counsellor's diary Cou\_029session recording Tda\_061the counsellor's diary 

What the records say

Open a record.

What you believe happened

Nothing has been contradicted yet.

Certainty100%

Records are illustrative, written for this piece — not transcribed from the game.

Start over 

## Chapter 3: Cognitive Friction

There's a text file flickering in the corner of the screen, accompanied by a sterile file directory structure, and audio logs laced with incomprehensible noise. The artwork that fills the screen in Serial Experiments Lain represent a world that is nothing like Mario's. Here, there are no lithe sprites running and jumping as you input commands. Instead, you're faced with *fragments of information*.

I am referring to the 1998 PlayStation game here, released exclusively in Japan - not the television anime that aired the same year. The game and the anime were conceived as parts of the same multimedia project: they share a world, a protagonist, and a set of ideas. But they do not ask us to experience the world in the same way.

![](https://storage.ghost.io/c/c0/e7/c0e78dd7-e7af-4603-8ab9-09f05cab06f3/content/images/2026/10/lain_cou011_3x_smooth-1.png)

Serial Experiments Lain. Source: LaunchBox.

The anime *shows us* Lain. The game *tasks us with searching for her*. And that difference truly matters. The game's approach is fascinating because it forces us to apply the principles of interactivity I articulated earlier to this case: does the dramatic reduction of physical input here (given that the player isn't moving a digital avatar around) mean that there is no *interactive dialogue*?

Here, the object of interaction isn't so much the player's fingers hitting buttons. It is, rather, the player's *understanding* of concepts that progressively build in response to the player.

You open one record. Lain describes something that might sound straightforward. Then you find and open another record. This time, the details no longer quite fit. A memory that, at one time, appeared straightforward now looks different when placed alongside another record. Was the first account incomplete? Was the second one unreliable? Was the apparent contradiction present in the two records themselves or, perhaps, it's Lain herself?

You open another file. Then another. The more information you gather, the less certain you become about what you thought you understood.

This is where Lain diverges even further from Mario.

In Mario, a failed jump communicated something meaningful and specific in the moment: you were too early or too late. The next attempt, which happens almost immediately, offers the chance to test your understanding.

In Lain, the "failure" is much harder to locate. You can't simply point to a single moment and declare, "That was the mistake." Instead, you begin to wonder whether your own interpretation was the mistake. Was this a record of something that actually happened? Or maybe Lain is misremembering something? 

Either way, the controller is not the primary participant in this dialogue. Your mind - your understanding - is where the action is happening. New information does not simply push the story forward in some passive sense from the player's perspective (perhaps in the way a traditional film might); rather, new information often challenges the understanding we, the players, have constructed in our minds up to that point.

There's a push and pull cycle here too:

**Information → Interpretation → Friction/Doubt → Re-exploration → Re-interpretation**

If Mario is "I act and the world responds" then Lain is "I perceive and my understanding changes". In the former case, the world responds to the player's body. In the latter case, information provided by the world changes the player's understanding of their own ideas.

**The World Answers You (Mario) vs The World Enters You (Lain).**

Importantly, the differences between these experiences are not a question of *which one is more interactive*. We're talking about the exact same push/pull idea. It's just that the push/pull is happening *in different places*.

The "Thousand Logs" in this story's title does not simply refer to the tragedy of a girl caught in a web of information. It also refers to the very structure of the experience itself: in order to approach and understand the existence of Lain, the player must gather countless fragmented records - logs - and find the connections between them.

One of the strangest aspects of this experience is that more information does not necessarily make the world clearer. Sometimes, the more we learn, the more we come to doubt our existing beliefs.

"Is this true?"

"Whose record is this?"

"Who is the *I* sitting in front of this screen?"

This is where the concept of cognitive friction becomes important. Cognitive friction doesn't refer to a deliberately inconvenient interface; it's the finely-tuned resistance that emerges between the information we're given by an artwork and what we believe we've understood about it. When a work does not immediately gratify us with clear answers, we stop, we interpret, we doubt...and we return to the world once again. The point at which we're forced to stop and think is precisely what I mean by cognitive friction.

Of course, friction is not inherently valuable. Being *difficult* or *inconvenient* is not the same thing as being forced to think deeply. What matters here is whether or not a work leaves room for players to *create meaning for themselves* rather than performing the act of thinking on their behalf.

What Lain demonstrates, in an extreme form, is that the value of interactivity cannot necessarily be measured by *how frequently the player operates something*. Sometimes, the player isn't moving a character; they're moving their own understanding. Each time that understanding changes, the world looks different.

## Chapter 4: Information Is Not Interaction

In November 1995, at Nintendo's annual trade show at Makuhari Messe in Chiba, Japan, a new generation of game hardware and cutting-edge graphics were drawing attention.

Amid all the excitement, Hiroshi Yamauchi, then president of Nintendo, stood before members of the press and the game industry and offered his thoughts on the essence of games:

**Hiroshi Yamauchi**President of Nintendo, November 1995

とにかく大容量を使って、音声と映像を垂れ流し、ゲームに迫力をつけようとするのです。しかしゲームというモノは、その内容が肝心なのです。音声や映像を垂れ流しても、愚作は愚作、駄作は駄作。おもしろくない、楽しくもないものは、ダメソフトなんです。

> In any case, they try to use large amounts of data, simply pouring out sound and images in an attempt to give games more impact. But when it comes to games, what matters is their content. A bad game is still a bad game, and a foolish game is still a foolish game, no matter how much sound and imagery you pour out. If it isn't interesting or fun, then it's simply bad software.

This was an era when high-capacity media such as CD-ROMs were opening up new possibilities for games. The ability to handle longer video, richer audio, and vastly greater amounts of data meant that games could represent more than ever before.

This was also the context in which Nintendo chose to continue with ROM cartridges for the Nintendo 64, even as Sony and Sega were moving toward CD-ROM-based systems. Yamauchi's argument about software quality therefore also served, at least in part, as a defense of Nintendo's own format choice in the face of criticism from an industry increasingly embracing CD-ROM.

At the same time, Yamauchi's words pointed to a simple fact: an increase in capacity or expressive power does not, by itself, guarantee that a game will become more interesting.

Looking back from 2026, however, his words can also be read as posing another question:

**Does an increase in the amount of information necessarily mean an increase in interaction?**

Of course, I am not suggesting that Yamauchi foresaw today's algorithms or AI in 1995\. Nor should we simply take whatever he meant at the time and retroactively translate it into the theoretical language of 2026\. But there is value in reading his criticism of using "large amounts of data" to simply "pour out sound and images" from where we stand today, especially given the amount of information that technology can process today, which has increased by orders of magnitude since 1995\. Images are higher resolution. Audio is spatial and richer. High-speed networks allow us to access information from around the world almost instantly. And today, algorithms can analyse our past behaviours and suggest what we should watch, listen to, or read next.

But this is where we need to pause for a moment and remember that *information is not interaction*.

The existence of a vast amount of information and the presence of a dialogue between ourselves and said information are two different things. No matter how much data a system can contain, no matter how spectacular its visuals or audio might be, none of it represents an increase in interaction if it asks nothing of the player (either their actions or interpretation).

This is where I return to the Lain example. Yes, it's an experience that places an enormous amount of data in front of players (in the form of audio, images, and text). But it isn't simply *content to be consumed*. It is, rather, deliberately designed to encourage a rich dialogue with players that demand their attention and active participation.

Let's consider how Yamauchi's words might be interpreted in today's media landscape. It's not that he was being prophetic, but rather, it's that the truth of his argument has remained so through time. And, in 2026, it's an argument that is thrown into far sharper relief.

It's not just that there's more information than ever available for our consumption, it's that today's media attempts to predict what we want and place the answer in front of us before we even asked for it. Before we've decided what to watch, the next video is queued up. Before we've decided what to listen to, a personalised playlist has been assembled. And, sometimes, before we've even found the words to express a question, AI has already generated an answer.

**What happens when the world answers us before we even ask a question?**

A demonstration that runs on its own: a question is typed out, and the answer appears before the question is finished. The exchanges are, in turn: "what should I watch tonight" answered with "Playing now."; "how do I say this without sounding" answered with "Here is the rest of your sentence."; and "what do I actually think about" answered with "Here is what you think."

You

Already answered

## Chapter 5: The World Answers Before You Ask

The "information being poured out" that Hiroshi Yamauchi worried about in 1995 has, by 2026, transformed into something else entirely. 

We are gradually reaching the point where we no longer need to actively search for information or even sift through enormous amounts of data to find the particular piece of content we want. Video platforms provide content that can be consumed "in the background" (where the assumption is that we'll primarily be focused on our phones or other devices). Recommendation algorithms present one option after another before we are consciously aware of what we want to watch next.

The arrival of generative AI has only accelerated these trends.

So, before we've had time to *form a question* \- let alone think about it or even begin searching for the words to express it - a system can predict what we are asking and generate an answer.

Let me return to the push/pull structure I've examined so far:

- **Mario:** **The World Answers You.** (The world responds to your own action.)
- **Lain:** **The World Enters You.** (Information from the world unsettles your understanding of yourself.)
- **Predictive Media:** **The World Answers Before You Ask.** (The world answers before you have even asked the question.)

This third horizon is where the relationship between humans and media begins to change.

With Mario, the player acts. The world responds. The player receives that response and chooses the next action.

With Lain, information enters the player from the world. The player interprets it, doubts it, and responds. The change of interpretation itself produces the next action.

Both of these cases contain that push/pull relationship.

In the algorithmic case, though, this relationship breaks down. An algorithm *predicts an action the human has not yet performed*, which means that it assumes ownership over at least some of that human action in advance.

I want to call this structure **delegation**.

This is not simply a question of convenience. It actually means that the very essence of human/media interaction is being delegated, at least in part, to an algorithm.

This is not simply a story about something becoming "more convenient."

It means that part of the push/pull that once existed between human beings and media is gradually being delegated to the system.

At first glance, it might seem that reducing the steps between our starting point and destination greatly increases convenience. And while this might be true on some level, there's more to the story. This is where games offer a fascinating counter example.

Consider Nintendo's Super Guide, the assistance feature Shigeru Miyamoto first described as "Demo Play". When a player gets stuck at some difficult section, the game can step in and show them how to get through it. This might initially look like the same kind of delegation I just described: player encounters a problem, game performs the difficult part, and the player no longer has to do it themselves.

But there's an important difference.

The idea here isn't to throw up our hands and say "You have failed too many times, so we will do it for you". Rather, the game can function almost like a friend who is better at the game: "Here. This is how you get through it. Why not try again?"

The player can then decide how to respond. Maybe they want to ignore the advice completely. Or maybe they want to take one small idea from it. In any case, the game doesn't demand that the player surrender to automation; it instead offers a form of guidance, leaving the player to respond in their own way. The game helps to unblock the immediate problem, but the controller can still ultimately return to the player. And that last point is the whole distinction. The help has to be proactively requested, it provides visible guidance, and hands control back to the player. The algorithmic example, by contrast, races ahead to choose and start playing the next video before any user choice is involved; it *evades* interactive conversation in favour of prioritising what it views as the *likely best decision*.

Nintendo's minimalist approach matters because, all the way through, it maintains the player/media interaction and it defers the decision-making to the player at critical points. Importantly, the player may not necessarily want the game to complete the tough challenge on their behalf in the first place. Maybe they were thinking: "Ah, I see what I did wrong. That was my mistake. I think I can do it this time, let me try again!"

In fact, this approach also illustrates that the player's motivation to act hasn't been replaced or weakened by the guidance: it has been *strengthened*. The player can turn the visual example into a new attempt, which might reduce just enough frustration to keep them in the game and, hopefully, to experience the satisfaction of successfully overcoming the challenge through their own work.

Remember, too, that this kind of guidance has been around for many years in different forms. Nintendo once offered something similar over the telephone through its Power Line - the Game Counselor hotline that Howard Phillips (the company's "Game Master" in America) helped establish - that let players call in and ask a counselor for advice when they became stuck.

Although this was an analogue system, the principle was the same: the idea was to reduce player frustration *without* reducing their *agency*. Sometimes, dialling the friction back just enough can give the player enough confidence to return to the challenge and overcome it themselves.

Only now do Yamauchi's words from 1995 begin to resonate from another angle. The problem of "information being poured out" was not simply about there being too much information; the deeper issue was that an increase in information *and what we actually do with it* are two different things. And in 2026, that problem has become even more complicated.

Now, the world answers before you ask. The distance between question and answer has become shorter and shorter. What happens when it becomes zero? Or, indeed, *less than zero* because player intent can be ascertained before the player is even consciously aware of it?

In this scenario, are we still in dialogue with the media? Or is the media in dialogue with itself - or with a synthetic copy of us - where we are relegated to observer-only status? When the media prompts with a question, who is actually answering?

## Epilogue: Who Is Answering?

Where should we go from here? Algorithms anticipate what we *might soon want* and place it in front of us. Generative A.I. does some thinking for us. Entertainment, in general, continues to peel away every last bit of friction.

This new *frictionless world* leaves us with the possibility that, over time, we may surrender our own agency around what we choose, think, and how we interact.

Let's not lose sight of this: The conclusion of this story isn't simply an appeal to nostalgia or the idea that "games were better in the past", nor is it a reactionary rejection of technological progress itself. What Mario taught us, by lending us his body, was the joy of a physical dialogue in which our own actions change the world.

**The World Answers You.**

And what Serial Experiments Lain confronted us with was the weight of cognitive friction: the need to face unsettling fragments of information where we reconstruct our own understanding with our own minds.

**The World Enters You.**

Whether the form of expression changes from pixel art to high-resolution computer graphics, or from static logs to predictive A.I., one question remains when we consider the relationship between human beings and media:

**Does the cycle of push/pull still remain?**

Great interactive media is not merely about presenting us with a beautiful world designed to astonish our eyes. Nor is it a convenient proxy that takes over every difficult act of thinking and choosing on our behalf. The media we should seek, at the very least, must leave room for us to give something back.

*A game should not merely show us a world. It should give us a world that we engage with, that "answers back".*

And that answer does not necessarily have to be comfortable (Mario can miss his jump, a Lain log can make understanding more difficult). What matters is not that the medium prioritises *completion*, but that it offers us the opportunity to be an active participant in the journey.

No matter how far A.I. advances, no matter how well systems can anticipate our questions (and answers), the thing we must preserve isn't so much the idea that we need to do everything with our own hands; it is that we always preserve that push/pull relationship where we retain our agency.

Sure, we can delegate calculations to machines. Algorithms can make personalised recommendations for us. A.I. can help us refine our grammar. I am not arguing against these things. But I believe the question we should always ask is whether or not we're allowing the push/pull to disappear.

Whether our reflexes, emotions, logic, memory - or all of the above - are being engaged, great interactive media will always present worlds that respect player agency.